# Balance of Payments for US Not Zero

I am following the lecture notes here

http://coin.wne.uw.edu.pl/brokicki/wsp_images/oem___set_1.pdf

which says balance of payments for a country, is zero

CA + KA + FA + SD + RES = 0

The variables are current account, capital account, financial account, statistical discrepancy, reserves. I attempted to verify this balance using data from Fred.

import pandas as pd, datetime, os
from pandas_datareader import data
start=datetime.datetime(1970, 1, 1)
end=datetime.datetime(2018, 5, 1)
df = data.DataReader(['BPCFTT01USQ637N','BPBLTD01USQ636N'], 'fred', start, end)
df.columns = ['finacct','goodacct']
df.to_csv('trade.csv')
import pandas as pd
df = pd.read_csv('trade.csv',parse_dates=['DATE'])
df = df.set_index('DATE')
df = df.dropna(axis=0)
df['total'] = (df.finacct + df.goodacct) / 1e9
df['interpolate'] = pd.rolling_mean(df.total,window=10)
df[['total','interpolate']].plot()


In the code above I compute the sum of Capital Accounts and Financial Accounts, and Current Account Balance: Total Trade of Goods. I also calculate a running average on this sum. The plot looks like

The sum is not zero, the interpolation sort of hovers around -50 billion dollars. My question is can this be considered zero since US GDP > 15 trillion dollars, much higher than this discrepancy, or should I look for another time series to plug this hole?

Thanks,